Spaces:
Sleeping
Sleeping
File size: 14,277 Bytes
7e3b585 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 | import json
import numpy as np
import faiss
from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity
from typing import List, Dict, Any, Tuple
import os
import pickle
class SpaceKnowledgeBase:
def __init__(self, data_dir: str = "data"):
self.data_dir = data_dir
self.model = SentenceTransformer('all-MiniLM-L6-v2')
self.index = None
self.documents = []
self.embeddings = None
self.index_path = "knowledge_base_index.faiss"
self.docs_path = "knowledge_base_docs.pkl"
# Load or create knowledge base
self._load_or_create_index()
def _load_json_files(self) -> List[Dict[str, Any]]:
"""Load all JSON data files and extract documents"""
documents = []
# File mappings for different data types
data_files = {
'space_terminology.json': self._process_terminology,
'space_agencies.json': self._process_agencies,
'planets.json': self._process_planets,
'rockets.json': self._process_rockets,
'astronauts.json': self._process_astronauts,
'telescopes.json': self._process_telescopes,
'space_museams.json': self._process_museums,
'notable_peoples.json': self._process_notable_people
}
for filename, processor in data_files.items():
file_path = os.path.join(self.data_dir, filename)
if os.path.exists(file_path):
try:
with open(file_path, 'r', encoding='utf-8') as f:
data = json.load(f)
documents.extend(processor(data))
print(f"Loaded {filename}: {len(processor(data))} documents")
except Exception as e:
print(f"Error loading {filename}: {e}")
return documents
def _process_terminology(self, data: Dict) -> List[Dict[str, Any]]:
"""Process space terminology data"""
docs = []
for term in data.get('space_terms', []):
doc = {
'id': f"term_{term.get('id', '')}",
'type': 'terminology',
'title': term.get('term', ''),
'content': f"{term.get('term', '')}. {term.get('short_description', '')} {term.get('detailed_description', '')}",
'category': term.get('category', ''),
'metadata': {
'category': term.get('category', ''),
'difficulty': term.get('difficulty_level', ''),
'related_terms': term.get('related_terms', [])
}
}
docs.append(doc)
return docs
def _process_agencies(self, data: Dict) -> List[Dict[str, Any]]:
"""Process space agencies data"""
docs = []
for agency in data.get('space_agencies', []):
doc = {
'id': f"agency_{agency.get('id', '')}",
'type': 'agency',
'title': agency.get('name', ''),
'content': f"{agency.get('full_name', '')}. {agency.get('description', '')} Founded: {agency.get('founded', '')}. Country: {agency.get('country', '')}",
'category': agency.get('type', ''),
'metadata': {
'country': agency.get('country', ''),
'founded': agency.get('founded', ''),
'type': agency.get('type', ''),
'headquarters': agency.get('headquarters', ''),
'budget': agency.get('annual_budget', '')
}
}
docs.append(doc)
return docs
def _process_planets(self, data: Dict) -> List[Dict[str, Any]]:
"""Process planets data"""
docs = []
for planet in data.get('planets', []):
doc = {
'id': f"planet_{planet.get('id', '')}",
'type': 'planet',
'title': planet.get('name', ''),
'content': f"{planet.get('name', '')}. {planet.get('description', '')} Distance from Sun: {planet.get('distance_from_sun', '')}. Type: {planet.get('type', '')}",
'category': planet.get('type', ''),
'metadata': {
'type': planet.get('type', ''),
'distance_from_sun': planet.get('distance_from_sun', ''),
'diameter': planet.get('diameter', ''),
'moons': planet.get('moons', ''),
'key_features': planet.get('key_features', [])
}
}
docs.append(doc)
return docs
def _process_rockets(self, data: Dict) -> List[Dict[str, Any]]:
"""Process rockets data"""
docs = []
for rocket in data.get('rockets', []):
doc = {
'id': f"rocket_{rocket.get('id', '')}",
'type': 'rocket',
'title': rocket.get('name', ''),
'content': f"{rocket.get('name', '')}. {rocket.get('description', '')} First flight: {rocket.get('first_flight_year', '')}. Purpose: {rocket.get('purpose', '')}",
'category': rocket.get('type', ''),
'metadata': {
'country_of_origin': rocket.get('country_of_origin', ''),
'operator': rocket.get('operator', ''),
'first_flight_year': rocket.get('first_flight_year', ''),
'payload_capacity': rocket.get('capacity_payload_kg', ''),
'active': rocket.get('active', False)
}
}
docs.append(doc)
return docs
def _process_astronauts(self, data: Dict) -> List[Dict[str, Any]]:
"""Process astronauts data"""
docs = []
for astronaut in data.get('astronauts', []):
doc = {
'id': f"astronaut_{astronaut.get('id', '')}",
'type': 'astronaut',
'title': astronaut.get('name', ''),
'content': f"{astronaut.get('name', '')}. {astronaut.get('description', '')} Agency: {astronaut.get('agency', '')}. Country: {astronaut.get('country', '')}",
'category': astronaut.get('type', ''),
'metadata': {
'country': astronaut.get('country', ''),
'agency': astronaut.get('agency', ''),
'birth_year': astronaut.get('birth_year', ''),
'missions_count': astronaut.get('missions_count', ''),
'achievements': astronaut.get('achievements', [])
}
}
docs.append(doc)
return docs
def _process_telescopes(self, data: Dict) -> List[Dict[str, Any]]:
"""Process telescopes data"""
docs = []
for telescope in data.get('telescopes', []):
doc = {
'id': f"telescope_{telescope.get('id', '')}",
'type': 'telescope',
'title': telescope.get('name', ''),
'content': f"{telescope.get('name', '')}. {telescope.get('description', '')} Type: {telescope.get('type', '')}. Status: {telescope.get('status', '')}",
'category': telescope.get('type', ''),
'metadata': {
'type': telescope.get('type', ''),
'country': telescope.get('country', ''),
'agency': telescope.get('agency', ''),
'year': telescope.get('year', ''),
'status': telescope.get('status', '')
}
}
docs.append(doc)
return docs
def _process_museums(self, data: Dict) -> List[Dict[str, Any]]:
"""Process space museums data"""
docs = []
for museum in data.get('space_museums', []):
doc = {
'id': f"museum_{museum.get('name', '').replace(' ', '_').lower()}",
'type': 'museum',
'title': museum.get('name', ''),
'content': f"{museum.get('name', '')}. {museum.get('famous_for', '')} Located in {museum.get('city_or_region', '')}, {museum.get('country', '')}. {museum.get('additional_info', '')}",
'category': 'space_museum',
'metadata': {
'country': museum.get('country', ''),
'city_or_region': museum.get('city_or_region', ''),
'famous_for': museum.get('famous_for', ''),
'established_year': museum.get('established_year', ''),
'annual_visitors': museum.get('annual_visitors', ''),
'additional_info': museum.get('additional_info', '')
}
}
docs.append(doc)
return docs
def _process_notable_people(self, data: Dict) -> List[Dict[str, Any]]:
"""Process notable people data"""
docs = []
for person in data.get('notable_space_contributors', []):
doc = {
'id': f"person_{person.get('name', '').replace(' ', '_').lower()}",
'type': 'notable_person',
'title': person.get('name', ''),
'content': f"{person.get('name', '')}. {person.get('contribution', '')} Known for: {person.get('known_for', '')}. Country: {person.get('country', '')}",
'category': 'space_pioneer',
'metadata': {
'country': person.get('country', ''),
'contribution': person.get('contribution', ''),
'known_for': person.get('known_for', ''),
'birth_date': person.get('birth_date', ''),
'death_date': person.get('death_date', ''),
'awards': person.get('awards', [])
}
}
docs.append(doc)
return docs
def _create_embeddings(self, documents: List[Dict[str, Any]]) -> np.ndarray:
"""Create embeddings for documents"""
texts = [f"{doc['title']} {doc['content']}" for doc in documents]
embeddings = self.model.encode(texts, show_progress_bar=True)
return embeddings
def _create_faiss_index(self, embeddings: np.ndarray) -> faiss.IndexFlatIP:
"""Create FAISS index for cosine similarity search"""
# Normalize embeddings for cosine similarity
faiss.normalize_L2(embeddings)
# Create index
dimension = embeddings.shape[1]
index = faiss.IndexFlatIP(dimension) # Inner product for cosine similarity
index.add(embeddings)
return index
def _load_or_create_index(self):
"""Load existing index or create new one"""
if os.path.exists(self.index_path) and os.path.exists(self.docs_path):
try:
# Load existing index and documents
self.index = faiss.read_index(self.index_path)
with open(self.docs_path, 'rb') as f:
self.documents = pickle.load(f)
print(f"Loaded existing knowledge base with {len(self.documents)} documents")
return
except Exception as e:
print(f"Error loading existing index: {e}")
# Create new index
print("Creating new knowledge base...")
self.documents = self._load_json_files()
if not self.documents:
print("No documents found!")
return
self.embeddings = self._create_embeddings(self.documents)
self.index = self._create_faiss_index(self.embeddings)
# Save index and documents
faiss.write_index(self.index, self.index_path)
with open(self.docs_path, 'wb') as f:
pickle.dump(self.documents, f)
print(f"Created knowledge base with {len(self.documents)} documents")
def search(self, query: str, top_k: int = 5) -> List[Tuple[Dict[str, Any], float]]:
"""Search for relevant documents using vector similarity"""
if not self.index or not self.documents:
return []
# Create query embedding
query_embedding = self.model.encode([query])
faiss.normalize_L2(query_embedding)
# Search
scores, indices = self.index.search(query_embedding, top_k)
# Return results with documents and scores
results = []
for i, (score, idx) in enumerate(zip(scores[0], indices[0])):
if idx >= 0 and idx < len(self.documents):
results.append((self.documents[idx], float(score)))
return results
def get_context_for_query(self, query: str, max_context_length: int = 2000) -> str:
"""Get relevant context for a query to use with LLM"""
results = self.search(query, top_k=5)
context_parts = []
current_length = 0
for doc, score in results:
doc_text = f"**{doc['type'].title()}: {doc['title']}**\n{doc['content']}\n"
if current_length + len(doc_text) > max_context_length:
break
context_parts.append(doc_text)
current_length += len(doc_text)
return "\n".join(context_parts)
def force_regenerate(self):
"""Force regeneration of the knowledge base"""
print("🔄 Force regenerating knowledge base...")
# Remove existing files
if os.path.exists(self.index_path):
os.remove(self.index_path)
if os.path.exists(self.docs_path):
os.remove(self.docs_path)
# Recreate
self._load_or_create_index()
print(f"✅ Knowledge base regenerated with {len(self.documents)} documents")
|